[Imageworld] LabelMe: An open-source web-based tool for image annotation

• Previous message: esann: "CFP: ESANN'2006 European Symposium on Artificial Neural Networks"
• Messages sorted by: [ date ] [ thread ] [ subject ] [ author ] [ attachment ]

Research on object recognition requires large amounts of images for
training and testing algorithms. The goal of this project is to
provide researchers with a
simple annotation tool and the possibility of having immediate access to a
large amount of data. LabelMe is not only a database of labeled images, but also
a web-based tool for easily creating and sharing more labeled data.

The LabelMe database contains thousands of labeled images. To get the
database, we only ask you to contribute to it by annotating some images.
After labeling some images, you will be granted access to the full
database, and that will include your latest annotations too.

This project provides a web-based annotation tool that makes it easy to annotate
images, and to instantly share such annotations with the community. This tool,
plus an initial set of 10,000 images (3000 of which have been labeled), can be
found at

       http://www.csail.mit.edu/~brussell/research/LabelMe/intro.html

Bryan Russell, MIT
Antonio Torralba, MIT
Kevin Murphy, UBC
William Freeman, MIT
_______________________________________________
Imageworld mailing list
Imageworld@itu.dk
http://mailman.itu.dk/mailman/listinfo/imageworld

• Previous message: esann: "CFP: ESANN'2006 European Symposium on Artificial Neural Networks"
• Messages sorted by: [ date ] [ thread ] [ subject ] [ author ] [ attachment ]